"The one downside to loading this model from TFHub is that the structure of internal keras layers is not restored. So it's more difficult to inspect or modify the model. The `TransformerEncoder` model is now a single layer:"
"The one downside to loading this model from TFHub is that the structure of internal keras layers is not restored. So it's more difficult to inspect or modify the model. The `BertEncoder` model is now a single layer:"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {},
"colab_type": "code",
"id": "pD71dnvhM2QS"
},
"outputs": [],
...
...
@@ -1536,8 +1387,6 @@
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {},
"colab_type": "code",
"id": "nLZD-isBzNKi"
},
"outputs": [],
...
...
@@ -1552,7 +1401,6 @@
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "ZxSqH0dNAgXV"
},
"source": [
...
...
@@ -1560,13 +1408,12 @@
"\n",
"### Low level model building\n",
"\n",
"If you need a more control over the construction of the model it's worth noting that the `classifier_model` function used earlier is really just a thin wrapper over the `nlp.modeling.networks.TransformerEncoder` and `nlp.modeling.models.BertClassifier` classes. Just remember that if you start modifying the architecture it may not be correct or possible to reload the pre-trained checkpoint so you'll need to retrain from scratch."
"If you need a more control over the construction of the model it's worth noting that the `classifier_model` function used earlier is really just a thin wrapper over the `nlp.modeling.networks.BertEncoder` and `nlp.modeling.models.BertClassifier` classes. Just remember that if you start modifying the architecture it may not be correct or possible to reload the pre-trained checkpoint so you'll need to retrain from scratch."
]
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "0cgABEwDj06P"
},
"source": [
...
...
@@ -1577,43 +1424,38 @@
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {},
"colab_type": "code",
"id": "5r_yqhBFSVEM"
},
"outputs": [],
"source": [
"transformer_config = config_dict.copy()\n",
"bert_encoder_config = config_dict.copy()\n",
"\n",
"# You need to rename a few fields to make this work:\n",
" \u003ca target=\"_blank\" href=\"https://www.tensorflow.org/official_models/nlp/customize_encoder\"\u003e\u003cimg src=\"https://www.tensorflow.org/images/tf_logo_32px.png\" /\u003eView on TensorFlow.org\u003c/a\u003e\n",
" \u003c/td\u003e\n",
" \u003ctd\u003e\n",
" \u003ca target=\"_blank\" href=\"https://colab.research.google.com/github/tensorflow/models/blob/master/official/colab/nlp/customize_encoder.ipynb\"\u003e\u003cimg src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" /\u003eRun in Google Colab\u003c/a\u003e\n",
" \u003c/td\u003e\n",
" \u003ctd\u003e\n",
" \u003ca target=\"_blank\" href=\"https://github.com/tensorflow/models/blob/master/official/colab/nlp/customize_encoder.ipynb\"\u003e\u003cimg src=\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\" /\u003eView source on GitHub\u003c/a\u003e\n",
" <a target=\"_blank\" href=\"https://www.tensorflow.org/official_models/nlp/customize_encoder\"><img src=\"https://www.tensorflow.org/images/tf_logo_32px.png\" />View on TensorFlow.org</a>\n",
" </td>\n",
" <td>\n",
" <a target=\"_blank\" href=\"https://colab.research.google.com/github/tensorflow/models/blob/master/official/colab/nlp/customize_encoder.ipynb\"><img src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" />Run in Google Colab</a>\n",
" </td>\n",
" <td>\n",
" <a target=\"_blank\" href=\"https://github.com/tensorflow/models/blob/master/official/colab/nlp/customize_encoder.ipynb\"><img src=\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\" />View source on GitHub</a>\n",
"# Assert that the variable `rezero_alpha` from ReZeroTransformer exists.\n",
"assert 'rezero_alpha' in ''.join([x.name for x in classifier_model.trainable_weights])"
]
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "6PMHFdvnxvR0"
},
"source": [
...
...
@@ -470,7 +447,6 @@
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "D6FejlgwyAy_"
},
"source": [
...
...
@@ -485,13 +461,9 @@
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {},
"colab_type": "code",
"id": "nFrSMrZuyNeQ"
},
"outputs": [],
"source": [
"# Use TalkingHeadsAttention\n",
"hidden_cfg = dict(default_hidden_cfg)\n",
...
...
@@ -508,12 +480,13 @@
"\n",
"# Assert that the variable `pre_softmax_weight` from TalkingHeadsAttention exists.\n",
"assert 'pre_softmax_weight' in ''.join([x.name for x in classifier_model.trainable_weights])"
]
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "kuEJcTyByVvI"
},
"source": [
...
...
@@ -528,13 +501,9 @@
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"colab": {},
"colab_type": "code",
"id": "XAbKy_l4y_-i"
},
"outputs": [],
"source": [
"# Use TalkingHeadsAttention\n",
"hidden_cfg = dict(default_hidden_cfg)\n",
...
...
@@ -551,12 +520,13 @@
"\n",
"# Assert that the variable `gate` from GatedFeedforward exists.\n",
"assert 'gate' in ''.join([x.name for x in classifier_model.trainable_weights])"
]
],
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {
"colab_type": "text",
"id": "a_8NWUhkzeAq"
},
"source": [
...
...
@@ -564,29 +534,26 @@
"\n",
"Finally, you could also build a new encoder using building blocks in the modeling library.\n",
"\n",
"See [AlbertTransformerEncoder](https://github.com/tensorflow/models/blob/master/official/nlp/modeling/networks/albert_transformer_encoder.py) as an example:\n"
"See [AlbertEncoder](https://github.com/tensorflow/models/blob/master/official/nlp/modeling/networks/albert_encoder.py) as an example:\n"
" \u003ca target=\"_blank\" href=\"https://www.tensorflow.org/official_models/nlp/nlp_modeling_library_intro\"\u003e\u003cimg src=\"https://www.tensorflow.org/images/tf_logo_32px.png\" /\u003eView on TensorFlow.org\u003c/a\u003e\n",
" \u003c/td\u003e\n",
" \u003ctd\u003e\n",
" \u003ca target=\"_blank\" href=\"https://colab.research.google.com/github/tensorflow/models/blob/master/official/colab/nlp/nlp_modeling_library_intro.ipynb\"\u003e\u003cimg src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" /\u003eRun in Google Colab\u003c/a\u003e\n",
" \u003c/td\u003e\n",
" \u003ctd\u003e\n",
" \u003ca target=\"_blank\" href=\"https://github.com/tensorflow/models/blob/master/official/colab/nlp/nlp_modeling_library_intro.ipynb\"\u003e\u003cimg src=\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\" /\u003eView source on GitHub\u003c/a\u003e\n",
" <a target=\"_blank\" href=\"https://www.tensorflow.org/official_models/nlp/nlp_modeling_library_intro\"><img src=\"https://www.tensorflow.org/images/tf_logo_32px.png\" />View on TensorFlow.org</a>\n",
" </td>\n",
" <td>\n",
" <a target=\"_blank\" href=\"https://colab.research.google.com/github/tensorflow/models/blob/master/official/colab/nlp/nlp_modeling_library_intro.ipynb\"><img src=\"https://www.tensorflow.org/images/colab_logo_32px.png\" />Run in Google Colab</a>\n",
" </td>\n",
" <td>\n",
" <a target=\"_blank\" href=\"https://github.com/tensorflow/models/blob/master/official/colab/nlp/nlp_modeling_library_intro.ipynb\"><img src=\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\" />View source on GitHub</a>\n",
"### Build a `BertPretrainer` model wrapping `TransformerEncoder`\n",
"### Build a `BertPretrainer` model wrapping `BertEncoder`\n",
"\n",
"The [TransformerEncoder](https://github.com/tensorflow/models/blob/master/official/nlp/modeling/networks/transformer_encoder.py) implements the Transformer-based encoder as described in [BERT paper](https://arxiv.org/abs/1810.04805). It includes the embedding lookups and transformer layers, but not the masked language model or classification task networks.\n",
"The [BertEncoder](https://github.com/tensorflow/models/blob/master/official/nlp/modeling/networks/bert_encoder.py) implements the Transformer-based encoder as described in [BERT paper](https://arxiv.org/abs/1810.04805). It includes the embedding lookups and transformer layers, but not the masked language model or classification task networks.\n",
"\n",
"The [BertPretrainer](https://github.com/tensorflow/models/blob/master/official/nlp/modeling/models/bert_pretrainer.py) allows a user to pass in a transformer stack, and instantiates the masked language model and classification networks that are used to create the training objectives."
"### Build a BertSpanLabeler wrapping TransformerEncoder\n",
"### Build a BertSpanLabeler wrapping BertEncoder\n",
"\n",
"[BertSpanLabeler](https://github.com/tensorflow/models/blob/master/official/nlp/modeling/models/bert_span_labeler.py) implements a simple single-span start-end predictor (that is, a model that predicts two values: a start token index and an end token index), suitable for SQuAD-style tasks.\n",
"\n",
"Note that `BertSpanLabeler` wraps a `TransformerEncoder`, the weights of which can be restored from the above pretraining model.\n"
"Note that `BertSpanLabeler` wraps a `BertEncoder`, the weights of which can be restored from the above pretraining model.\n"
"### Build a BertClassifier model wrapping TransformerEncoder\n",
"### Build a BertClassifier model wrapping BertEncoder\n",
"\n",
"[BertClassifier](https://github.com/tensorflow/models/blob/master/official/nlp/modeling/models/bert_classifier.py) implements a [CLS] token classification model containing a single classification head."
"With the `loss`, you can optimize the model. Please see [run_classifier.py](https://github.com/tensorflow/models/blob/master/official/nlp/bert/run_classifier.py) or the colab [fine_tuning_bert.ipynb](https://github.com/tensorflow/models/blob/master/official/colab/fine_tuning_bert.ipynb) for the full example."
]
}
],
"metadata": {
"colab": {
"collapsed_sections": [],
"name": "Introduction to the TensorFlow Models NLP library",
